There is strong evidence that galactose, a sugar derived from milk, is involved in the etiology of epithelial ovarian cancer, through ovarian damage, infertility and hypergonadotrophism. Galactose is an important component of the glycolipids and glycoproteins on cell surfaces, structures regulating onco-fetal development and cell-to-cell recognition, and determining blood-group substances. Integrity of these compounds, and their type, defines susceptibility to infections and governs neoplastic progression. Some 15% of women carry abnormally functioning gene variants for GALT, a key enzyme metabolizing galactose. This genetic epidemiology study will enroll 320 patients with ovarian cancer, 166 of their sisters and 576 genetically unrelated in-laws, at Memorial Sloan-Kettering Cancer Center and New York Hospital. Subjects will be interviewed and blood will be taken to test for abnormal GALT gene alleles and other biomarkers. Live cultured lymphocytes and sera will be frozen and stored for future genetic and virologic studies. Conventional case-control analyses will explore how the GALT gene interacts with dietary milk products, and with other genes and conventional epidemiologic factors, to alter the risk of epithelial ovarian cancer. Specific questions to be answered include: Is the risk of ovarian cancer increased in women carrying gene alleles that cause the GALT enzyme to function abnormally? Does consumption of galactose increase the risk? Are women with normal GALT alleles able to consume milk products freely without increasing risk? Is having "lactose intolerance" protective? Do oral contraceptives reduce the risk of ovarian cancer specifically by modifying the influence of the GALT enzyme or the diet? Other genes to be tested as potential risk factors are those that control the structure of certain glycolipids and glycoproteins: those determined by alleles of the P, Lewis, Secretor, ABH(O) and MNS blood-group substances. Also, the proportions of certain glycolipids in cultured lymphocytes will be tested as potential intermediate biomarkers of disturbances in glycolipid synthesis; if they are found to be altered by GALT abnormalities their relationship to ovarian cancer and diet will be explored.